Error in Saving in old version

[Gliby Anna]

Dear All

Please help me solve this error

data cannot be saved in old .dta format
Data contain strL or str#, #>244, and prior releases of Stata would not know how to
process these variables. Either drop the variables or use recast with the force
option to change them to str244.

[daniel klein]

The error message already tells you your options. What exactly is the problem? What exactly is the "solution" that you are looking for?

Do you have difficulties identifying the respective variables? Type

describe

and look for strL or str#, where #>244 in the storage type (2nd) column; these are the variables that you cannot use in older Stata versions.

Besides dropping or recasting those variables, you could also try to split them. If you want that, please say so and we can help you with code that does this. Include examples when possible.

Best
Daniel

[Gliby Anna]

Hi Daniel

Yes I want to know how can I identify the variable. Then maybe split them or drop

Thank you

[Gliby Anna]

Dear Danial

Thanks I was able to do it

[daniel klein]

A more systematic way of identifying the variables is

ds , has(type 245/2045 strL)

You could then refer to r(varlist) where the respective variables are stored. If you go for one of the brute-force approaches that Stata suggests, i.e. drop or recast, you could automate this. Here is an example

ds , has(type 245/2045 strL)
drop `r(varlist)'

Best
Daniel